1Some versions of this model use the price level instead of the inflation rate to make the modelmore consistent with its microeconomics counterpart. Using the inflation rate, as is done here, producesresults that are a little more realistic.

Figure 2.1

CHAPTER 2THE AGGREGATE SUPPLY - AGGREGATE DEMAND MODEL

The first formal macroeconomics model introduced by the text is called the Aggregate Supply - Aggregate DemandModel, which will hereafter be referred to as the AS/AD model. The AS/AD model is useful for evaluating factors andconditions which effect the level of Real Gross Domestic Product (GDP adjusted for inflation) and the level of inflation.

The model is an aggregation of the elementary microeconomic supply-and-demand model discussed in the previouschapter. Like the microeconomic model, the AS/AD model is a comparative statics model. The model's insights, therefore, areobtained by identifying and initial equilibrium condition, then "shocking" the model by charging one or more of the parameters,then evaluating the resulting new equilibrium.

Introduction to the Aggregate Supply/Aggregate Demand Model

Now that the structure and use of a basic supply-and-demand model has been reviewed, it is time to introduce theAggregate Supply - Aggregate Demand (AS/AD)model. This model is a mere aggregation of themicroeconomic model. Instead of the quantity ofoutput of a single industry, this model representsthe quantity of output of an entire economy (or, inother words, national production).

Refer to Figure 2.1 for an example of theAS/AD model. As can be seen, two variables arerepresented by the model. The quantity variableon the horizontal axis is now represented by realgross domestic product (Y). This is the measureof the true value of annual national production,and is adjusted for inflation. The level of priceinflation is represented on the upright axis. Asuitable economic statistic for this value would bethe rate of inflation as determined by the ImplicitPrice Deflator, the value used to compute realGNP from nominal, inflated GNP.1

The equilibrium level of real nationaloutput (real GDP) is determined by theinteraction of the AS and AD curves. Thisequilibrium also determines the national inflationrate.

The Aggregate Demand (AD) curve has its traditional negative slope. This implies that, given any amount of nominalincome, purchasers will be able to buy more real goods at lower prices than they would at higher prices. Aggregate demandis simply the total of all levels of spending in the national income accounts; consumption, investment, government purchases,and net exports.

The Capacity Utilization Rate and the AS Curve

Look carefully at the slope of the Aggregate Supply (AS) curve. As can be seen, it is relatively flat for a considerablepart of its length, then turns sharply upward, turning nearly vertical past some point. The relatively flat region of the AS curveis the non-inflationary region, whereas the nearly vertical portion is the inflationary region of the AS curve.

Page 2

2It is worth noting that policy-makers at the Federal Reserve System, the nation's monetaryauthority, watch this statistic very carefully.

The sharp bend in the curve can be associated with an economic statistic called the Capacity Utilization Rate. Thisstatistic is a crude measure of the percent of capacity being used by the nation's manufacturers. Theoretically, if an industryis running at 100 percent capacity, all available productive assets are being used. Although certain industries can run at 100%capacity or even above for brief periods of time, the national average, which this statistic represents, is seldom above 90%, andin a typical non-recession year will be between 80% and 85%. Since the 1960s, during recessions it has dropped below 80%.In the recession that ended in 1982, it fell to below 70%.

Why might the economy begin to experience symptoms of inflation when the capacity utilization rate goes above 85%?This figure is a national average, and when the average is this high, certain "heated" industries are likely to be well above 90%.These particular industries are likely to be facing serious pressures on costs as they encounter resource bottlenecks or laborshortages. They may be paying workers overtime, paying higher prices for their raw materials, or encountering costlyproduction inefficiencies. These rising costs eventually translate into higher prices for finished goods, and as this becomes truefor more and more industries, price increases become more widespread.

By no means is the capacity utilization rate a perfect indicator of the economy's movement from a non-inflationaryenvironment to an inflationary environment. For one thing, it includes a measure of only manufacturing capacity and henceexcludes the important service sector. It could not forewarn, for example, of inflation in the medical services sector.Nonetheless, it can serve as a crude first approximation of the inflationary turning point. If the capacity utilization rate isaround 80% or below, inflation is not very likely, where as when this statistic is above 85%, a continued growth in aggregatedemand is much more likely to be associated with rising inflation.2

Factors Effecting Aggregate Supply and Aggregate Demand

Like the microeconomic supply-and-demand model, changes in equilibria in the AS/AD model are caused by changesin the variables that effect supply and demand. Refer to Figure 2.2. Again, the variables that are likely to effect supply ordemand are listed. The presumed direction ofinfluence is shown with a (+) or (--) sign, aswas done with the microeconomics model.

Again, the relationship between AS,AD and price are represented by the slope ofthe AS and AD curves; changes in all othervariables cause the curves to shift right orleft.

A review of the list shows someoverlap or redundancy. For example, bothinterest rates and credit availability arerelated of course, and one might be used tothe exclusion of the other. Despite the factthat these are related, there is a differencebetween them. For example, credit extendedby credit cards became more readily availableto consumers in the late 1970s andth roughou t the 1980s becausecomputerization lowered transaction costs.This is an institutional reason for creditavailability and would be reflected in a modelconcerned with showing the effects of thisinstitutional change.

Factors that Effect Aggregate Supply and Aggregate Demand

Aggregate Demand Aggregate Supply

1. Income2. Wealth3. Population4. Interest rates5. Credit availability6. Government demand7. Taxation8. Foreign demand9. Investment10. Expectations (a) Inflationary (b) Income (c) Wealth (d) Interest rate

(+)(+)(+)(–)(+)(+)(–)(+)(+)

(+)(+)(+)(+)

1. Costs (a) Labor (wages) (b) Resource2. Investment (prior)3. Productivity4. Interest rates5. Credit availability6. Foreign supply7. Expectations (a) Profits (b) Inflationary (c) Interest rate8. Taxation

(–)

(+)(+)(–)(+)(–)

(+)(?)(?)(–)

(+): An increase in this factor causes the curve to shift right.(–): An increase in this factor causes the curve to shift left.

Figure 2.2

Page 3

3There are considerable tax advantages of using home equity loans to encourage this activity aswell. Interest paid on home equity loans are deductible from taxable income, reducing the consumer'sincome tax burden.

4Unfortunately, nothing is ever quite so simple as this. Some theories claim, for example, that evena balanced budget stimulates the economy. The reader might refer to any discussion of the balanced budgetmultiplier in any introductory economic textbook.

In the case of deficit spending, there may be some crowding out of private spending because of theimpact in the finance markets. This is discussed in the next chapter.

The list shown is not all-inclusive, but is certainly a good starting point. Some of the factors and their respective signsare self-evident and will not, therefore, be discussed further. Others, though, require some comment.

Wealth and Wealth Expectations

It should be obvious that income would be strongly correlated with aggregate demand since that is primarily howdemand is financed, but the effect of wealth and expectations of future wealth (and for that matter income expectations) isperhaps not so obvious. The argument here is that consumer perceptions of their own wealth, which is linked to income but notthe same thing, and expectations of future wealth will effect their consumption behavior. A reasonable measure of wealth wouldcertainly include consumer savings (including bank deposits and ownership of liquid financial assets, such as stocks and mutualfunds) but would also include more abstract and less liquid categories of wealth, such as home equity. Consumers, in otherwords, with considerable equity (the difference between the market value and what is owed on the loan balance) in their homeshave more latitude for spending than consumers who do not. This is especially true if they expect that equity to grow.

One might ask how consumers can increase their spending if the source of wealth is illiquid (not easily converted tocash) or is based upon expected rather than realized income and wealth. One source of potential demand is obvious; savings.The general availability of consumer credit is another. Consumers expecting an increase in income can finance purchasesimmediately with credit cards or installment credit (such as that used to purchase automobiles) and make payments from futureincome. Likewise, consumers living in states like New York and California, having experienced substantial increases in homeequity (at least up until 1989) perceived that their wealth had grown and could raise their standard-of-living with the popularhome equity loans.3

In summary, when savings are ample or credit is readily available, spending is not directly restricted by immediate,realized income, and aggregate demand can respond to consumer perceptions of wealth, expected income, or expected wealth.

Government Demand and Taxation

Government demand in this context is the equivalent of Government Purchases of Goods and Services in the nationalincome account. The treatment of transfer payments like Social Security, which merely pass income from one private party,the taxpayer, to another, the recipient, must depend on how item 7, taxation, is treated. If transfer payments are excluded fromgovernment demand, then so must the proportion of taxation used to finance transfer payments.

Taxation has a negative sign for aggregate demand for two reasons: (1) it reduces disposable income for consumersand, (2) because it lowers business profits, it lowers any investment that might have been financed by those profits. Becausefrom the perspective of business taxation is yet another cost of doing business, at least to some degree, it has the same effectupon aggregate supply as other cost categories. An increase in taxation tends to reduce aggregate supply.

On net then, government spending tends to stimulate demand whereas taxes tend to retard demand. Therefore, agovernment running a balanced budget would have a roughly neutral effect on aggregate demand and a government running abudget deficit (where expenditures exceed revenues) would have a stimulating effect.4

Page 4

5The incidence of various types of taxes upon costs is a very complicated issue and is normallyconsidered in the context of microeconomics. Sales taxes, for example, will have a different impact thancorporate income taxes. Any detailed discussion of this is beyond the scope of this text. Here it is beingassumed that business taxes have the same general effect upon supply as costs.

6This explanation requires one more qualification. Since the inflation rate is being represented onthe vertical axis, it is more accurate to say that an increase in the rate of cost inflation will shift theaggregate supply curve to the left.

Figure 2.3

Costs and Productivity

An increase in any category of costs will tend to shift the aggregate supply curve upwards. This might include costs of rawmaterials, transportation or energy costs, labor costs, or even business taxes.5

To help understand the impact of costsupon aggregate supply, refer to Figure 2.3.That diagram shows the aggregate supply curveshifting left because producers are facing highercosts. To understand the justification for this,ask the question, "Choosing any level ofproduction, such as Yx, at what price willproducers be willing to supply that amountgiven rising costs?". That answer must be, "Ata higher price". Generalized, the leftward shiftof the aggregate supply curve implies thatproducers will be willing to supply any givenlevel of output, such as Yx, only at higherprices when faced with rising costs. The failureto increase prices to cover rising costs willotherwise drive some producers out ofbusiness.6

Consider now the impact of an increasein economic productivity. This effect is shownin Figure 2.4. Generally, anything whichincreases the level of output given the level ofresources used is increasing productivity.Labor productivity is increased when more output is obtained from the same or less labor. For example, if the automobileindustry is able to manufacture more automobiles with few workers, this is said to reflect an increase in labor productivity.

In a mature industrial economy, an increase in productivity is primarily a technological phenomenon. Applied researchand development, resulting in automation, new technology and better engineering will raise productivity. This is especially trueof labor productivity, where automation allows more output per worker. Nor are productivity increases restricted to themanufacturing sector. Significant gains have been made in the service sector by the widespread use of, for example, computers,telecommunications equipment, and modern office equipment. The productivity of secretaries has been increased by word-processing computers and high-speed, efficient duplicators. Salesmen can be made more efficient with FAX machines andcellular telephones. Because of computerization, services ranging from retail trade to banking have seen significant increasesin productivity through the 1980s.

By inspection it can be seen that increases in productivity shift the aggregate supply curve to the right. The implicationis obvious and significant. It implies that technological improvement, better engineering, adequate levels of research anddevelopment, and anything else that might raise productivity tends to allow an increase in the standard of living, at least as

Page 5

7Note that this argument depends upon the aggregate demand curve being located in theinflationary region of the aggregate supply curve. If the aggregate demand curve were located well back onthe non-inflationary range of the aggregate supply curve, when the capacity utilization rate is, say, below70%, this argument would not obtain. Wage increases or other stimulants to aggregate demand may verywell be justified in a severe recession or depression. Under such circumstances the economy has a lot oflatitude for non-inflationary growth.

Figure 2.4

Figure 2.5 Graph (a)

measured by real national output, while at thesame time working against inflationary pressures.Productivity gains provide the ultimate long-runsalvation for a mature economy.

Consider, for example, the impact ofdemands for higher wages in an economy that isnot experiencing productivity gains compared toone that is. Workers have the right to expect anincrease in their standard of living over time, andthey will try to raise their incomes through wageincreases.

Wage increases will raise income and,hence, cause aggregate demand to rise (shiftingthe aggregate demand curve to the right in themodel). Because wages are part of the coststructure of doing business, the aggregate supplycurve should shift right. As graph (a) of Figure2.5 demonstrates, if the wage increases occurwhen the economy is running at near-full-capacity (i.e. in the inflationary range of theaggregate supply curve) and there is noproductivity gain, the result is almost pureinflation with no substantial increase in thestandard of living. Worker's expectations ofhigher real income are then disappointed byinflation. Nominal wages rise, but purchasingpower is eroded by general inflation.7

However, if productivity gains have beensufficient enough to offset the impact of risingwage costs, the aggregate supply curve will shiftright, as shown in graph (b) of Figure 2.5. Inthis case the economy experiences non-inflationary growth, and worker expectations forhigher real income (and a higher standard ofliving) are satisfied.

A rather important economic lesson iscongealed in this example. Aspirations for ahigher standard of living are common in a matureeconomy, and can manifest themselves in waysranging from agitation for higher wages to theextensive use of credit. Despite the avenuechosen and the best of intentions, though, theaggregate desire to raise the standard of living by

Page 6

Figure 2.5 Graph (b)

"demand-side" means will only result in inflationand unless the economy is expanding its means toproduce higher levels of output. As will be seenlater, this tension does play a partial role inmodern business cycles.

Although most of the emphasis in thisdiscussion has been on the productivity variable,another in the list, previous investment, plays anearly identical role. Obviously productivity andprevious levels of business investment arestrongly related.

One final comment should be made. Tosay that productivity gains are necessary does notimply that economic growth must be associatedwith the ever-larger production of resource-intensive commodities. In an age of resourcedepletion and serious environmental concerns,future productivity gains will (or should) be ofthe sort that enhances the level of economicservice provided by the economy while usingfewer natural resources. Progress in this area isalready well underway. Automobiles are betterand safer now, for example, than two decades ago, yet they weigh less, burn less fuel, and waste fewer resources. Growth, inother words, can be accomplished with resource savings. That is what productivity measures.

Expectations and the Aggregate Supply Curve

When referring to the list of factors that effect the aggregate supply curve, it can be seen that both inflationary andinterest rate expectations have an ambiguous effect upon the aggregate supply curve, hence the (?) sign associated with them.The effect of these expectations upon production decisions will probably depend upon the type of business. Businesses withlarge inventories, for example, might accelerate production in an environment of inflationary expectations (or acceleratepurchases if running large resource inventories for their production). Business with low or no inventories but experiencing laborshortages may, in contrast, cut back production (or avoid expansion) if they expect their labor costs to rise. In the case ofinterest-rate expectations, real estate construction might slow if interest rates were expected to rise (because high interest ratesdiscourage final sales), but in capital goods formation, businesses might accelerate their debt-financed expansion plans in orderto finance at current lower rates rather than anticipated higher rates. Therefore the net effect of these expectations uponaggregate supply is ambiguous.

Applications of the AS/AD Model toBusiness Cycle Analysis

Now that the AS/AD model has been developed, it can be applied to certain economic questions that are asked in theevaluations of business cycles. The examples provided below are typical. More will be provided in their proper contextthroughout the text.

In this section, for each issue considered, a question will be posed and then addressed using the AS/AD model. Againthe reader is reminded that a model will not provide the final, decisive answer on a complex question. The model will insteadprovide perspective and insight.

Here are the questions that will be considered in this section:1. How is a business cycle represented by this model?2. Will a stimulation to aggregate demand cause growth or inflation?3. How can a business cycle result from disappointed consumer expectations?

Page 7

Figure 2.6

4. How can stagflation be explained by this model?This series of questions is not all-inclusive. Certain policy questions can be addressed by this model, but those will be

presented in the chapters on policy.

1. How is the business cycle represented by this model?

Refer to Figure 2.6. The business cyclein the AS/AD model is shown by the movementof equilibria over time as the AS curve, the ADcurve, or both shift over time. Figure 2.6 showsonly a plotting of equilibrium points. The ASand AD curves are not shown (only theirintersections). It should be remembered thataggregate demand is usually more volatile thanaggregate supply, so most of the shift inequilibria would like come from movements ofthe AD curve.

Economic expansion is shown as amovement of the equilibrium from left to rightbecause such a movement represents growth inreal GNP. A recession, in contrast, is shown asa movement in the opposite direction, from rightto left, because that represents a contraction ofreal GNP (real GNP is falling, which implies thegrowth rate is negative). An upward movementof the equilibrium represents an increase in therate of inflation, whereas a movement downrepresents a decline in the rate of inflation.

Consider point 'a' in Figure 2.6 as the beginning point for a recovery, near the trough of the recession. The movementfrom 'a' to 'b' represents the normal and healthy phase of recovery and expansion. There is substantial growth in real GNP anda moderate increase in price inflation. The latter condition need not necessarily be present - it is conceivable that the rate ofinflation could remain constant or even decline under unusual circumstances. What is shown is the more typical case. Thismovement from 'a' to 'b' would normally be mostly due to a shift right in AD as demand grows (and would be reflected in anincrease in the capacity utilization rate) but there would also be some shift to the right by the AS curve as investment andproductivity gains increased the economy's capacity to provide goods.

If the late phase of the expansion moved into a strongly inflationary phase (as often the case as not), this would bereflected by a movement in equilibria such as that from 'b' to 'c'. In this late phase of the expansion, there is less real growthand higher inflation rates. Obviously the AD curve has moved into the inflationary range of the AS curve. At point 'c'maximum real growth has been reached and the cycle is at its peak.

What happens next depends upon whether inflation continues for awhile after the economy begins to recess (the typicalcase in the modern era) or whether it immediately begins to abate. The former case is represented in Figure 2.6 as the movementfrom 'c' to 'd'. There is a recession because real GNP is falling, but the inflation rate continues to rise. Finally, past point 'd'the inflation rate abates as the recession continues. In a mild recession, the final "leftward" resting point of the equilibrium willbe far to the right of the original point 'a' and probably point 'b'. Only in a severe depression would real GNP fall back as faras the starting point of the previous recovery. Remember, recessions are relatively mild compared to expansions.

The path shown is meant to be representative. Each historical cycle would be a little different. Certainly in all businesscycles there is first a movement in the equilibrium from left to right during the expansion, then a movement from right to leftduring the contraction. The mix of inflation with the cycle varies from cycle to cycle, however which would alter the shape ofthis path from one cycle to the next.

2. Will a stimulation of Aggregate Demand Cause Growth or Inflation?

Page 8

Figure 2.7 Graph (a)

Figure 2.7 Graph (b)

As stated above, historical inflation ratesare not perfectly consistent over the phases ofdifferent cycles. Inflation, nonetheless, isintricately related to the dynamics of the cycle.Because aggregate demand is relatively volatilecompared to aggregate supply and is so easilystimulated by policy, the question posed above isstrongly related to business cycle dynamics.

An inspection of the list of factors thatcan effect aggregate demand in Figure 2.2expose many candidates for a stimulation toaggregate demand. In the area of policy, forexample, either an increase in governmentspending or a decrease in taxes can shift the ADcurve right. Expansionary monetary policy,increasing credit availability and reducinginterest rates, can do the same. So can a growthin consumer confidence, represented by anincrease in income expectations. In any of thesecases, the AD curve would shift sharply right.

For the sake of discussion, assume thestimulation came from expansionary monetarypolicy. With such a stimulus, interest rates would drop and both credit and money supply growth rates would surge. Aggregatedemand would respond by shifting right.

Would this necessarily cause aninflation? Simplistic theories about theconnection of monetary growth rates to inflationrates say yes - inflation is caused by "too muchmoney chasing too few goods". Therefore, if themoney supply growth rate is high, inflation willfollow.

Complicated economic relationships areseldom well represented by simple ideas orslogans, however. Figure 2.7 shows why this istrue when considering the relationship betweenexpansionary monetary policy (and high creditand money supply growth rates) and inflation.

If a strong stimulus to aggregate demand,such as an expansionary monetary policy, isapplied near the trough of a recession, as isshown in graph (a) of Figure 2.7, there is plentyof room for non-inflationary real growth withvery little inflation. If the starting point, in otherwords, is well back on the non-inflationary rangeof the AS curve, inflation is not likely to be aproblem.

In the real economy this might imply that the capacity utilization rate, at the starting point of the stimulus, is unusuallylow - perhaps below 70%. Likewise the unemployment rate is likely to be high and resources available. There would not bemany bottlenecks, labor shortages or severe cost constraints in the economy. With their excess capacity, manufacturers couldsatisfy the resulting surge in sales by utilizing existing capacity. Inflationary pressures would not emerge.

Page 9

Figure 2.8

If on the other hand, the same monetary stimulus were applied well into the expansionary phase of the cycle, as is shownin graph (b) of Figure 2.7, where excess capacity is diminished, resource and labor constraints are emerging, or, in other words,when the economy is moving into the inflationary region of the AS curve, inflation will be the result. The stronger the stimulus,the higher will be the inflation rate and the less will be the growth of real GNP.

Furthermore, if the expansion is inflationary, as shown in graph (b), the result obtained there is not the end of the story.The movement into the inflationary range of the AS curve implies that in the real economy, inflation would gradually beexperienced by the general population. According to the theory of adaptive inflationary expectations, robust inflationaryexpectations would be formed as consumers experienced more inflation. This might transpire over a few months. This wouldlikely provide a secondary stimulus to aggregate demand (and might possibly effect aggregate supply as well, contracting it).The final result is shown in Figure 2.8.

The initial inflation caused by themonetary stimulus is shown as the movement inthe AD curve from AD to AD2. It results in anincrease in the inflation rate from P1 to P2. Asinflationary expectations begin to be formed fromthis experience, however, a secondary, additionalstimulus effects aggregate demand, moving theAD curve out to AD3. This continued surge ininflation, from P2 to P3, reflects the effect ofinflationary expectations. Such expectations,therefore, have the effect of compounding theinflation.

This particular scenario, as it wasdescribed, assumed the formation of adaptiveexpectations, which are formed from experience.Had national expectations been assumed(probably not appropriate in this context), thefinal effect would have been the same, but theresult would have transpired in far less time.Instead of learning from inflation, the inflationaryeffects of the monetary stimulation would havebeen deduced directly and immediately, and theexpectations would have been formed and demand shifted over a very short period of time.

These are two general lessons to be learned from this application. First, the impact of a demand-stimulating policy (orany other condition that stimulates aggregate demand) will depend upon the economic context in which the stimulation takesplace. To be more particular, if the stimulation occurs deep in a recession, the effect is almost entirely helpful, whereas in theheated phase of the cycle, it is harmful.

Second, the dynamics of an inflationary phase are complex because of the role played by expectations. Expectationstend to compound economic problems, as they have here, making solutions more elusive. This example provides one of thereasons why the Federal Reserve System, the nation's monetary authority, considers it wiser strategy to try to prevent aninflation, which might involve erring on the side of caution, rather than to let the inflationary dynamics begin and then try tocure the inflation.

3. How can a Business Cycle Arise from Disappointed Consumer Expectations?

As described before, the formation of expectations can strongly effect the dynamics of a business cycle. Expectationscan both initiate a cyclical trend or can amplify or mitigate the amplitude of the cycle, making a recession worse or milder,depending upon the types of expectations and how they are formed. Without question expectations will effect the timing of acycle.

Page 10

8There is some evidence from preliminary data that the recession beginning in 1990/91 followed,roughly at least, the scenario described below.

Figure 2.9

Figure 2.9 provides an example. In this case, the formation of consumer expectations of rising income and wealthprovide the dynamics of the cycle.8

Suppose in the final months of the expansion demand growth is very high for traditional reasons - all sectors arespending freely. This is shown in graph (a). It is during this phase that the economy and incomes are growing very rapidly,

Page 11

perhaps above 5% in real terms. Consumers slowly develop adaptive expectations that their incomes will continue to rise atthese high rates. Prices are also beginning to rise. Suppose the price increases are not evenly spread and prices for homes arerising higher than prices on average. Consumers who own homes see their equity grow (which is perceived as wealth) and thiscontributes to the formation of higher wealth expectations.

In graph (b), both income and wealth expectations are now robust, and consumers begin to borrow heavily to financemany of their purchases, especially of consumer durable goods, home improvements, vacations, and similar discretionaryexpenditures. In some cases they borrow against the equity in their homes. As shown, the economy becomes more heated.Fewer of the gains, however, are in real income. The inflation rate rises.

If consumer expectations have been unrealistic, their debt/income ratios will be rising over this phase of the cycle. Thismeans, of course, that consumer debt is outpacing their income, and the burden of servicing that debt, normally through monthlypayments, takes a growing percentage of disposable income. This is shown in graph (d), where the rates of debt-to-income ischarted over the cycle. When the economic equilibrium moved from 'b' to 'c' in graph (b), the debt-to-income ratio moved from'b' to 'c' in graph (d).

With their income expectations disappointed and heavily indebted, consumers begin to feel burdened by their debt,especially when they see that their incomes are not rising fast enough to allow them to easily service their debt. Their equity(hence wealth) may still be rising, but this does not provide them with the cashflow to service debt. Therefore, they try to reducetheir debt by cutting back on spending. If this is an endemic problem in the economy, demand can fall enough to initiate arecession, as is shown in graph (c) by the movement of the AD curve from AD3 to AD4 and the equilibrium from point 'c' topoint 'd'.

Unfortunately, this is not the end of the story. As the rate of inflation falls, wealth expectations are also disappointed,since equity growth either stabilizes or actually falls. This further discourages spending.

More important, even through consumers are trying to rectify their debt-to-income ratios by reducing debt throughreducing spending, it is ironically possible in this kind of environment for the debt-to-income ratio to continue to rise, contraryto consumer desires. This is because the fall in income due to the recession is greater than the reduction of debt that caused therecession. Mathematically, for the ratio of debt to income, although the numerator (debt) is falling, the denominator (income)is falling faster, causing the ratio to rise. This unwanted development is shown in graph (d) as the movement in the debt/incomeratio from 'c' to 'd'. Again, this is happening during the recession. Therefore, there is a secondary effect on spending, and therecession becomes deeper, moving from equilibrium 'd' to 'e'. At some point the recession stabilizes (perhaps for some otherreason, such as anti-recession policy), income stabilizes, the debt-to-income ratio stabilizes (graph (d), point 'e') or even drops,and the retraction stops.

The leverage in this recession is caused in part because when consumers try to reduce their debt/income ratios theyinadvertently increase it. This paradox of results being different from intentions, commonly found in economics, might best beexplained by the example of a single family. Suppose in a family both husband and wife work and the husband occasionallyworks overtime. Family income rises and their home equity rises. Feeling their new prosperity, the couple make homeimprovements, buy a new auto, some furnishings, and finance a vacation partly with debt, including a home equity loan. Theyeventually discover their debt burden is far higher than they anticipated and they have trouble making payments. Perhaps thehusband's overtime is reduced. They decide to cutback sharply on spending to pay off some of their debt. If this problem is,however, endemic, and many consumers are doing it, a recession results. The effect filters back to this family. Overtime iseliminated and perhaps one or both are temporarily unemployed. Certainly there is a loss of income. Well into the recessionthe family realizes, much to their grief, the debt burden is heavier than ever, and this is true for millions of consumers.

Why Expectations are Sometimes Wrong

The previous example depended upon the theory of adaptive expectations for the explanation. Expectations were slowto form and were learned from experience. Consumers formed expectations of rising incomes upon their recent experiences ofincome growth. This is a somewhat "impressionistic" view of the formation of expectations.

Critical to the dynamics of the cycle scenario just described was the point that consumer expectations were toooptimistic - their expected future wealth and income did not materialize. This introduces the question, "How could such a largegroup of consumers be so incorrect?"

Figure 2.10 provides the answer to this question. If the theory of adaptive expectations is correct, if the consumersexperience a few months or years depending upon the variable) of high growth rates of some key variable, they will begin to

Page 12

Figure 2.10

Figure 2.11

expect more of the same. Figure 2.10 isgeneralized to refer to any cyclical variable thatmight give rise to adaptive expectations. Supposefor the sake of discussion, the variable in questionis consumer income. As the illustration shows, ifconsumers experience a few years of high incomegrowth, represented by the movement from 'a' attime 't0' to 'b' at time 't1' (say five years later),they will expect income to continue to grow atapproximately the same rate, expecting it to be atlevel 'ce' at time 't2' (a few years in the future).They then adjust their economic behavior to theseexpectations. If the growth rate of income thentapers off for any reason, and 'ca' is the actuallevel of income realized at time 't2', then consumerexpectations are disappointed.

As shown in the bottom of Figure 2.10,expectations can work both ways. If incomegrowth, to use that example, has been flat for afew years, then expectations will be weak, asshown. Then if income begins to rise, it isperceived as a windfall.

Obviously, the shorter the time horizonfor the formation of adaptive expectations, themore strongly expectations are likely to beinfluenced by recent experience and short-runtrends. Consequently, short-sighted or myopicexpectations are likely to add volatility to normalcyclical events.

4. How Can Stagflation by Explained bythis Model?

Stagflation is the combination ofinflation and recession and is, hence, the worstpossible combination in the cycle. At leastduring an ordinary recession, prices tend to fall.However in some recessions inflation peaks nearthe trough of the cycle.

The combination of inflation andrecession is relatively easy to explain with theAS/AD model however. Refer to Figure 2.11.The AS curve is shown shifting backwards, fromAS, to AS2, causing the economic equilibrium tomove from 'a' to 'b'. As can be seen, the newequilibrium is associated with a higher inflationrate and a reduction in real output (by definition,a recession). This combination is calledstagflation.

The shift backward in the AS curvewould be regarded as a rare and unusual event.

Page 13

As stated previously, the AD can be regarded as relatively volatile, but the AS curve is much more stable. Normally it driftsto the right over time as investment and productivity expand the economy's capacity to supply goods and services. Any sizeablecontraction would be unusual.

A huge short-run impact upon costs, however, can have this effect. The OPEC petroleum embargo in the 1970s, forexample, may have had this effect. The first full embargo was in 1973 and the second in 1979, both coming just prior torecessions. The embargo restricted supplies of petroleum coming into the United States, and prices for energy, gasoline andother petroleum fuels, synthetic made from petroleum, and so forth, shot up sharply. Petroleum and energy is such an integralpart of the economy that the cost structure was "shocked" to some degree.

Any severe resource constraint can have this effect. If there is, for example, a critical shortage of skilled labor in thefuture, pushing up wage costs, stagflation might return and would be explained in much the same way as the explanationprovided in Figure 2.11. At a regional rather than national level, such as in the Western States, a severe shortage of water mighthave the same effect.

A Cautionary Note

Other applications of the AS/AD model will be shown in later chapters. The reader is reminded that the model can'tpossibly provide a full explanation for cycle phenomena. It is merely a template which frames up the discussion and offersinsight. The application on stagflation, for example, hardly provide an exhaustive explanation of the complicated causes of therecessions in the 1970s and early 1980s, for example. It does suggest, however, that the petroleum embargo contributed to theunusual character of those recessions. So long as the model's limits are recognized and results interpreted as suggestive ratherthan definitive, the AS/AD model can be a useful tool of analysis.

©Gary R. Evans, 1999. This version of this chapter may be used for educational purposes without permission of the author. Instructors are encouraged to use this material inclass assignments either by allowing students to download the material from a web site or by photocopying the material. Both options are free subject only to the limitations statedhere. The material may be photocopied for reproduction under the following three conditions: (1) this copyright statement must remain attached to all copies; (2) students and otherusers may be charged only for the direct cost of reproduction; (3) this material is used for non-profit educational purposes only. The material may not be sold for a profit withoutexplicit written consent of Gary R. Evans.

As a courtesy but not as an obligation, the author would like to informed by email at evans@hmc.edu of any use of this chapter in a classroom context, and would appreciatecomments suggestions for improving the content.